Open Models Almanac
Methodspost-trainingreinforcement learning algorithm

single mention · post-training

Domain-specialized RL experts

Trains separate experts for broad domains, with each domain covering a range of subtasks and reasoning-effort levels.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See reinforcement learning algorithm for the methods filed beside it.

Also called scaling RL across three broad domains.

Filed under post-training › reinforcement learning algorithm